Federal authorities discovered $1 billion in fraud in one year thanks to artificial intelligence

New York (CNN) – The federal government’s commitment to using artificial intelligence to combat financial crimes appears to be paying off.

Machine learning AI helped the US Treasury sift through massive amounts of data and uncover $1 billion worth of check fraud in fiscal year 2024 alone, according to new estimates first shared with CNN. That’s nearly three times what the Treasury recovered in the previous fiscal year.

“It’s been really transformative,” Renata Miskel, a senior Treasury Department official, told CNN in a phone interview.
“Leveraging data has improved our ability to detect and prevent fraud,” Miskel said.

The Treasury Department credited artificial intelligence with helping officials prevent and recover more than $4 billion in fraud in fiscal year 2024, a six-fold increase from the previous year.

US authorities quietly began using artificial intelligence to detect financial crimes in late 2022, taking a cue from what many banks and credit card companies are already doing to stop criminals.

The goal is to protect taxpayer money from fraud, which has surged during the COVID-19 pandemic as the federal government rushed to disburse emergency aid to consumers and businesses.

The Treasury is certainly not using generative AI, the kind that has wowed users of OpenAI’s ChatGPT and Google’s Gemini by generating images, creating song lyrics, and answering complex questions (although it still sometimes struggles with Simple inquiries).

Instead, fraud detection efforts rely on machine learning, a subset of artificial intelligence that excels at analyzing large amounts of data and making decisions and predictions based on what it has learned.

AI can be very useful in combating financial crime, as it can comb through endless streams of data and discover subtle patterns, all in a fraction of the time it takes a human to do so. Experts say that once trained, advanced AI models can detect suspicious transactions in just fractions of a second.

“Fraudsters are very good at hiding. They are secretly trying to game the system,” says Miskel. “AI and leveraging data helps us find these hidden patterns and anomalies and work to prevent them.”

This is especially critical for the Treasury Department, which is among the largest, if not the largest, payers on the planet.
Each year, the Treasury makes about 1.4 billion payments worth nearly $7 trillion to 100 million people. It is responsible for sending everything from Social Security and Medicaid payments to federal workers’ payrolls, tax refunds and stimulus checks.

This critical role makes the Treasury Department a prime target for fraudsters trying to steal taxpayer money.

Last year, the I.R.S advertisement Which deployed artificial intelligence to detect tax fraudsters by examining large and complex returns from hedge funds, law firms and others.

Online payment fraud is expected to exceed $362 billion by 2028, according to estimates from Juniper Research.

And part of this fraud is driven by AI itself.

In a notorious case earlier this year, Hong Kong police said a financial employee was tricked into paying $25 million to fraudsters via a fake video.

US authorities have expressed concern that artificial intelligence will bring new risks to the financial system. Treasury Secretary Janet Yellen warned bankers in June that artificial intelligence in finance poses “significant risks.”

Top regulators, led by Yellen, classified AI late last year as an “emerging vulnerability” in the financial system.

Miskel stressed that while AI systems flag suspicious transactions, “a human is always aware” and federal agencies make the final decision on whether something constitutes fraud.

The Treasury Department’s use of artificial intelligence to combat financial crimes is just beginning.

Meskel indicated that the Treasury Department is studying how to adopt fraud detection methods deployed by major banks and credit card companies, but he did not want to go into details so as not to “alert bad actors.”

A Treasury Department spokesperson told CNN that the department is accelerating its work to improve the fraud detection tools available to federal and state-run programs. Officials are testing new data sources to better detect fraud and suspicious payments, and are working with state agencies to combat unemployment insurance fraud.